CLI Commands

Kubetorch offers a rich set of commands to offer you insight into running workloads at the individual and cluster level.

Because all workloads are running as services on Kubernetes, you can also use kubectl if you prefer to interact with your services.

kt billing

Show a summary of vCPU and GPU usage

kt check

Run a comprehensive health check for a deployed service

kt config

Get or set your username, namespace, and desired Kubetorch installation source URL for services launched with Kubetorch

kt dashboard

Open Kubetorch dashboard with Grafana

kt debug

Start an interactive debugging session on the pod, which will connect to the debug server inside the service

kt describe

Show basic info for calling the service from outside the cluster

kt deploy

Deploy a Python file or module to Runhouse. This will deploy all functions and modules decorated with @kt.compute in the file or module

kt list

See live services and resources that have been deployed with Kubetorch

kt logs <kt-service-name>

View the logs for a particular service

kt metrics

Open Grafana dashboard

kt queues

List pods that are currently queued

kt run

Build and deploy a kubetorch app that runs the provided CLI command. In order for the app to be deployed, the file being run must be a Python file specifying a kt.app construction at the top of the file.

kt ssh <kt-service-name>

Directly work on the remote compute by SSHing in (to the head node if distributed)

kt status

Load service status details

kt teardown <kt-service-name>

Tears down all related resources to a particular service